The reproductive epidemiology project emphasizes the development and application of new methods for measuring human reproductive damage. Reproductive outcomes include fertility, sub-clinical early fetal loss, spontaneous abortion, fetal growth, and birthweight. Each of these outcomes can be affected by environmental factors, and represents a possible endpoint for studying the effects of toxins on human reproduction. One major component of this project is the study of fertility. Time-to-pregnancy is being developed as a potentially useful and sensitive measure of fertility. For example, a strong relation between smoking and time to pregnancy has been documented. Another approach to the measurement of fertility involves the study of daily urine specimens from women who are trying to become pregnant. By comparing the ovarian hormones in cycles which results in pregnancy with hormones in infertile cycles, new methods may be developed for measuring fertility impairment. These methods may be applicable to women not actively trying to conceive. A second component of this project is the study of very early pregnancy loss. In a prospective study of 230 women who have stopped using birth control in order to become pregnant, daily urine specimens are being collected to be tested for evidence of pregnancy. This will provide the best available estimate of the extent of early pregnancy loss in humans. The risk of early loss will be studied in relation to common exposures in this population, such as use of alcohol, tobacco, caffeine beverages and medications. Work continues on the development of a new method for the analysis of birth weight and perinatal mortality.